Deep Learning

Anchor Boxes Data Pipeline References Mingxing Tan, Ruoming Pang, Quoc V. Le. EfficientDet: Scalable and Efficient Object Detection.

Differentiable Architecture Search $$ \begin{aligned} && \min_{\alpha} &&& \mathcal{L}_{val} (w^{\ast} (\alpha), \alpha) \\ && \text{s.t.} &&& w^{\ast} (\alpha) = \text{argmin}_{w} \space \mathcal{L}_{train} (w, \alpha) \\ \end{aligned} $$

The most successful ML projects in production (Tesla, iPhone, Amazon drones, Zipline) are where you own the entire stack. They iterate not just ML algorithms but also: 1) how to collect/label data, 2) infrastructure, 3) hardware ML models run on.

Introduction The Oil and Gas industry is one of the most lucrative industries that has a very high operating cost. Cutting costs is therefore a major priority when it comes to this business.

Note: Some details in the dataset have been obfuscated by the provider in order to maintain the data privacy. Modern recommendation systems are complicated. The Collaborative Filtering, which used to be the state-of-the-art approach during the Netflix Prize competition, is probably not at the forefront of the research or industry anymore.

1. Introduction 2. Model Implementation 3. Initial Results 3.1 Single Class Detection Simple Scenes Currently the model is trained on a single class: car. For the following result, green indicates the ground truth labels, and light blue indicates the predicted results.